Hermite Polynomials


by Domnu
Tags: hermite, polynomials
Domnu
Domnu is offline
#1
Aug15-08, 08:43 AM
P: 178
Since the Hermite Polynomials are orthogonal, could one state that they span all polynomials [tex]f[/tex] where [tex]f : R \rightarrow R[/tex]? This would be EXTREMELY useful for the harmonic oscillator potential in quantum mechanics...
Phys.Org News Partner Science news on Phys.org
Scientists pinpoint when harmless bacteria became flesh-eating monsters
Asian air pollution affect Pacific Ocean storms
Rocket leak delays space station delivery launch (Update)
Jimmy Snyder
Jimmy Snyder is offline
#2
Aug15-08, 01:14 PM
P: 2,163
Quote Quote by Domnu View Post
Since the Hermite Polynomials are orthogonal, could one state that they span all polynomials [tex]f[/tex] where [tex]f : R \rightarrow R[/tex]? This would be EXTREMELY useful for the harmonic oscillator potential in quantum mechanics...
No. If you think of the two vectors:
(1, 0, 0) and (0, 1, 0)
They are orthogonal to each other, but they don't span.
on the other hand, consider the three vectors:
(1, 0, 0), (1, 1, 0), (1, 1, 1)
They span, but they are not orthogonal to each other.
Therefore, knowing that the Hermite Polynomials are orthogonal is not enough to show that they span. You would need to prove that.
Avodyne
Avodyne is offline
#3
Aug15-08, 01:28 PM
Sci Advisor
P: 1,185
It doesn't follow from orthogonality alone, but it is nevertheless true.

Domnu
Domnu is offline
#4
Aug15-08, 02:00 PM
P: 178

Hermite Polynomials


Quote Quote by jimmysnyder View Post
No. If you think of the two vectors:
(1, 0, 0) and (0, 1, 0)
They are orthogonal to each other, but they don't span.
on the other hand, consider the three vectors:
(1, 0, 0), (1, 1, 0), (1, 1, 1)
They span, but they are not orthogonal to each other.
Therefore, knowing that the Hermite Polynomials are orthogonal is not enough to show that they span. You would need to prove that.
Fair enough... okay so I guess a better argument would be that the Hermite polynomials are infinite in number and are all orthogonal, so they span all functions?
Useful nucleus
Useful nucleus is offline
#5
Aug15-08, 02:07 PM
Useful nucleus's Avatar
P: 240
They are useful to express any function defined on the range on which they are orthogonal.
Other orthogonal polynomials will be useful to express functions on the range on which they are orthogonal (e.g. Legender polynomials on [-1,1].
morphism
morphism is offline
#6
Aug15-08, 02:21 PM
Sci Advisor
HW Helper
P: 2,020
Quote Quote by Domnu View Post
Fair enough... okay so I guess a better argument would be that the Hermite polynomials are infinite in number and are all orthogonal, so they span all functions?
No, that doesn't make any sense.

In what sense are you using the word "span"?
Domnu
Domnu is offline
#7
Aug15-08, 02:27 PM
P: 178
I'm using the word span in the sense that any polynomial of the form

[tex]f(\xi) = a_n \xi^n + a_{n-1} \xi^{n-1} + a_{n-2} \xi^{n-2} + \cdots + a_1 \xi + a_0[/tex]

can be written as a linear combination of [tex]H_0, H_1, \cdots, H_n[/tex]. I see I had a typo in my last post... I meant that the Hermite polynomials could span all polynomials (real).
Jimmy Snyder
Jimmy Snyder is offline
#8
Aug15-08, 05:24 PM
P: 2,163
Quote Quote by Domnu View Post
Fair enough... okay so I guess a better argument would be that the Hermite polynomials are infinite in number and are all orthogonal, so they span all functions?
No. For instance, the following are mutually orthogonal and infinite in number, but they don't span,
(0,1,0,0,0,...)
(0,0,1,0,0,...)
(0,0,0,1,0,...)
...
Dick
Dick is offline
#9
Aug15-08, 05:35 PM
Sci Advisor
HW Helper
Thanks
P: 25,158
Quote Quote by jimmysnyder View Post
No. For instance, the following are mutually orthogonal and infinite in number, but they don't span,
(0,1,0,0,0,...)
(0,0,1,0,0,...)
(0,0,0,1,0,...)
...
True. But it's pretty clear the Hermite polynomials span. H_n(x) contains a term containing x^n and none of the preceding H's do. You can formally prove it by induction, but it's actually pretty obvious if you imagine how you would go about expressing a given polynomial in terms of Hermite polynomials.
Jimmy Snyder
Jimmy Snyder is offline
#10
Aug15-08, 06:00 PM
P: 2,163
Quote Quote by Dick View Post
True. But it's pretty clear the Hermite polynomials span. H_n(x) contains a term containing x^n and none of the preceding H's do. You can formally prove it by induction, but it's actually pretty obvious if you imagine how you would go about expressing a given polynomial in terms of Hermite polynomials.
My guess is that Domnu would benefit by doing the induction.


Register to reply

Related Discussions
Hermite Polynomials. Calculus 4
Hermite Polynomials Advanced Physics Homework 0
Hermite Polynomials Advanced Physics Homework 0
Hermite Polynomials Calculus & Beyond Homework 2